long-context inference

ContrastKV: Robust KV Cache Eviction via Contrastive Signal Fusion for Multi-Query Generalization ACL 2026 Deep Kernel Fusion for Transformers ACL 2026 ECHO: Efficient KV Cache Offloading with Lossless Prefetching for Serving Native Sparse Attention LLMs OSDI 2026 Focus-dLLM: Accelerating Long-Context Diffusion LLM Inference via Confidence-Guided Context Focusing ACL 2026 HeteroCache: A Dynamic Retrieval Approach to Heterogeneous KV Cache Compression for Long-Context LLM Inference ACL 2026 LongSpec: Long-Context Lossless Speculative Decoding with Efficient Drafting and Verification ACL 2026 LoPT: Lossless Parallel Tokenization Acceleration for Long Context Inference of Large Language Model ACL 2026 No Buffer, No Bottleneck: Efficient Zero-Copy KV Cache Offloading for Long-Context LLMs OSDI 2026 RRAtention: Dynamic Block Sparse Attention via Per-Head Round-Robin Shifts for Long-Context Inference ACL 2026 S2O: Early Stopping for Sparse Attention via Online Permutation ACL 2026 SlimInfer: Accelerating Long-Context LLM Inference via Dynamic Token Pruning AAAI 2026 VecAttention: Vector-wise Sparse Attention for Accelerating Long Context Inference CVPR 2026 AdaSkip: Adaptive Sublayer Skipping for Accelerating Long-Context LLM Inference AAAI 2025 APB: Accelerating Distributed Long-Context Inference by Passing Compressed Context Blocks across GPUs ACL 2025 FIER: Fine-Grained and Efficient KV Cache Retrieval for Long-context LLM Inference EMNLP 2025 MadaKV: Adaptive Modality-Perception KV Cache Eviction for Efficient Multimodal Long-Context Inference ACL 2025 MEDA: Dynamic KV Cache Allocation for Efficient Multimodal Long-Context Inference NAACL 2025 MoQAE: Mixed-Precision Quantization for Long-Context LLM Inference via Mixture of Quantization-Aware Experts ACL 2025 QuantSpec: Self-Speculative Decoding with Hierarchical Quantized KV Cache ICML 2025 RAPID: Long-Context Inference with Retrieval-Augmented Speculative Decoding ICML 2025 TokenSelect: Efficient Long-Context Inference and Length Extrapolation for LLMs via Dynamic Token-Level KV Cache Selection EMNLP 2025 SnapKV: LLM Knows What You are Looking for Before Generation NeurIPS 2024